|Home | About | Journals | Submit | Contact Us | Français|
Because this was the first study using prebiotics for the prevention of atopic dermatitis, our statistical consultant proposed that the p value should be calculated “conservatively”. However, we agree that the single‐sided test is more appropriate, which indeed changes the p value from 0.014 (as given in the paper) to 0.008.
We were aware of the fact that the risks by parental history of various allergic diseases are not interchangeable, as described in the paper by Dold et al.3 We recorded the allergic diseases of the parents. There was no difference between the two groups with respect to the incidence of parental atopic dermatitis, allergic rhinitis, asthma, food allergy, drug allergy, etc.
Because both study formulas were supplemented with vitamin D (1.3 μg/100 ml), no extra supplementation was administered to the infants. Thus, the vitamin D intake was sufficient and similar in both groups. Consequently, we can not identify any bias caused either by inappropriate randomisation or by a different influence of vitamin D supplementation.
The policy of Macedonio Melloni Maternity Hospital in Milan (where the study was performed) strongly supports breast feeding. Thanks to a human milk bank, no baby receives any formula during their stay in hospital. After discharge, a free telephone line is available 24 h per day, 7 days per week, to deal with breastfeeding problems. Additionally, all infants after discharge from the hospital are checked in a special department dedicated to the promotion and support of breast feeding.4 This system was also used for the infants enrolled in this study. At discharge, the parents were asked to contact the hospital if they decided to start formula feeding; if they contacted the hospital they were then given information about the study. As demonstrated by the drop‐out analysis, due to the support of the hospital team, several mothers with infants randomised to one of the two formula groups were successfully encouraged to continue or to re‐establish breast feeding (17% and 16%, respectively). The result of the drop‐out analysis clearly indicates that the policy of the hospital did not change during the study period.
In the preceding year, 24.7% of the infants with a parental history of allergy/atopy developed atopic dermatitis during the first year of life, which is quite similar to the number we found in the control group of our study. Thus, there was no difference between the study period and the period with non‐observing conditions. Evaluation of data regarding the parental history of atopy in our hospital population revealed that the mothers tended to have a higher incidence of atopy compared to the fathers (a ratio of 10 to 7) which could explain the relatively high incidence of atopic dermatitis in the first year of life.
The study was planned at the end of 2002 and started in March 2003. At that time, registration was not essential for a clinical trial with an ingredient already on the market. Also, there were no reasons not to register because of confidentiality as the composition of the prebiotic mixture had already been published by our group.5 However, we agree that in view of current practice this study should be registered.
Competing interests: GB is working as Director of Infant Nutrition Research at Numico Research, Friedrichsdorf, Germany.